Understanding the events involved in the emergence of human pathogens from animal commensals/pathogens is important to our ability to confront their increasingly frequent outbreak. In an intriguing experiment of nature, the persistent commensal of a broad range of animals, Bordetella bronchiseptica, gave rise independently to two closely related human pathogens, B. pertussis and B. parapertussis. Both organisms are highly contagious, cause acute disease with high pathology and display strong resistance to antibodymediated clearance, characteristics not observed in their common progenitor. We have recently shown that B. pertussis uses Pertussis Toxin (PTx) to avoid rapid antibody-mediated clearance, allowing this organism to infect immune hosts; a defining characteristic of B. pertussis. Paradoxically, B. parapertussis causes the same disease in the same host, and contains the genes for PTx but does not express them. The historical view of the apparent convergent evolution towards the ability to infect humans is unable to explain this. Another paradox of the bordetellae is the very existence, and remarkable success, of these two very closely related organisms in the same host populations. Evolutionary theory would predict that two such closely related organisms should compete via immune-mediated pressures within the host population. In our efforts to explain these observations we have proposed that, in addition to selection for optimal interactions with their hosts, the evolution of the bordetellae has been largely shaped by immune-mediated competition between strains that inhabit the same host population. This model fits with the observations that B. bronchiseptica infects all the animals around us but only rarely infects humans, and often those that are immunodeficient. Our preliminary data show that B. bronchiseptica is indeed sensitive to immune-cross protection and rapidly eliminated from B. pertussis-immune animals. Thus the exclusion of B. bronchiseptica from most healthy humans is the result of the high level of B. pertussis-immunity. However B. pertussis and B. parapertussis, both closely related to B. bronchiseptica, have coexisted within the same human populations, often at the same time. Here we offer the central hypothesis that immune-mediated competition shapes the evolution of the bordetellae and, by extension, that B. parapertussis and B. pertussis are under strong selective pressure for the loss of cross reactive antigens. We use our experimental infection model to directly examine crossimmunity and the effects of expression of cross-reactive antigens. We also extend our published MLST based phylogeny to examine the sequence of a number of antigenic genes across a population of closely related Bordetella isolates from humans and animals. Our phylogenetic studies have identified sequence types that infect both humans and animals. Using our 454 machine we will efficiently sequence the entire genomes of these strains to examine the effects of immune-mediated pressures at the genome level. Finally, these data will parameterize models of the within- and between-host dynamics of these reemerging human pathogens.